Evidence for mixed feelings of happiness and sadness from brief moments in time

Abstract

Theorists disagree about whether valence is a basic building block of affective experience or whether the positive and negative substrates underlying valence are separable in experience. If positivity and negativity are separable in experience, people should be able to feel happy and sad at the same time. We addressed limitations of earlier evidence for mixed feelings by collecting moment-to-moment measures of happiness and sadness that required participants to monitor their feelings only occasionally. In Study 1, participants were occasionally cued to press one button if they felt happy and another if they felt sad. Participants spent more time reporting mixed feelings (i.e., simultaneously pressing both buttons) during bittersweet scenes than non-bittersweet scenes. In Study 2, participants reported their feelings only once. Participants spent more time reporting mixed feelings when cued during a bittersweet, as opposed to non-bittersweet, scene. These results extend earlier evidence that happiness and sadness can co-occur.

Keywords:

• Mixed feelings
• Mixed emotions
• Ambivalence
• Bittersweet
• Happiness
• Sadness

We thank Clyde Hendrick for suggesting that we conduct Study 1.

We thank Clyde Hendrick for suggesting that we conduct Study 1.

Notes

¹ We have referred to “mixed emotions” (Larsen & McGraw, 2011), but adopt the term “mixed feelings” to highlight that our focus is on the structure of what Russell and Carroll (1999) termed “core affect” (i.e., the conscious experience of affect), rather than the occurrence of discrete emotions.

² We use the terms “moment-to-moment” (Woltman Elpers, Wedel, & Pieters, 2003) and “real-time” as Larsen and McGraw (2011) used the terms “continuous” and “online,” respectively. We have revised our terminology for the sake of clarity.

³ A 2 (Order: Life is Beautiful first; The Garden of the Finzi-Continis first)×4 (Cue Type: control, happy, sad, bittersweet) ANOVA on the duration of happiness data revealed a significant Order×Cue Type interaction, F(3, 36)=4.13, p=.01. Among participants who watched The Garden of the Finzi-Continis first, all the significant differences shown in Table 1 were replicated. Participants who watched Life is Beautiful first spent more time reporting happiness during happy (Mdn=9.1 s), control (Mdn=9.3 s), and bittersweet (Mdn=6.2 s) cues than during sad cues (Mdn=0.0 s; all ps<.001), but spent no more time reporting happiness during happy cues than during control and bittersweet cues. Contrast effects may have produced these order effects. For instance, Life is Beautiful's tragic scenes may have made the uneventful clip from The Garden of the Finzi-Continis seem more pleasant.

⁴ All reported effects remained significant when data from the six participants (11%) who reported having previously seen Life is Beautiful were removed.